Occupancy Estimation Two Species - Error!

[GooseCG]

Hi all

Firstly, I am most definitely an amateur using this program! I studied it briefly during one module of my MSc and am now using the program as part of my data analysis in my thesis.

I am using the program to look at occupancy estimation for two different species. However, I get this error;

ERROR -- Encounter history was too short for the number of occasions specified.

I can't figure out what is wrong with my encounter history input files. In this particular file I have 120 encounters, below is just a small sample of that data;

/*B*/ 1101100010111001000100010100000011010000001001100100110001001010000100000001000001100111 1;
/*B2*/ 0000000000000000001000000000010000000001010000000001000100010000000000000010000000000000 2;

Does anyone have any idea why it may be misreading my encounter history, or the occasions I specify? Is it correct to have one species as 'B' and the other as 'B2' in the same file like this?

Apologies, I have not had much guidance on using this program and having trawled through various means of help I've decided it might be easier to directly ask those with experience!

Thanks for you help.

[markmiller]

I have a fair bit of experience with 2-species occupancy models, but I have always used SURVIV or base R for that. I may not have time to look into this much, but consider the following questions:

1. Which of the three (or more?) 2-species occupancy models in MARK are you using?

2. How many visits to a site were made?

3. How many primary periods were there? Where primary periods are 'years' or 'seasons'.

4. How many secondary periods were there? Where a secondary period includes visits within a year or within a season.

5. In my experience, typically the number of columns in a capture history is equal to the number of visits to a site which is equal to the number of visits in a secondary period times the number of primary periods. Although that might not be the case with every model in MARK.

Indeed, with two-species models I might expect the number of columns in a capture history to be twice the total number of visits to a site, as I explain below.

6. Is each visit to a site considered an encounter? You wrote that you have 120 encounters, but your capture histories only have 88 columns. I might expect the capture histories to have either 120 columns or 2 * 120 = 240 columns.

7. Sometimes a capture history might be structured like so:

11 01 10 00 4 ;

or

11011000 4 ;

or

AB 0B A0 00 4 ;

or

AB0BA000 4 ;


any of which might mean there was one primary period and four visits and:

1. on the 1st visit you detected both species A and B
2. on the 2nd visit you only detected species B
3. on the 3rd visit you only detected species A
4. on the 4th visit you did not detect either species

and this capture history was recorded at four different sites.

However, since I have never used the 2-species models in MARK I do not know whether the capture histories are structured that way.

These are just some basic suggestions that may or may not help. If I have time I might look into this more.

[GooseCG]

Hi Mark

Thank you for taking the time for explain this. I think I know where I might be going wrong - I had originally created two different encounter files for each species but it would seem that they need to be in one?

I shall have a play around and see what happens.

Thanks again for your help.